A State Graph-Based Improved Framework for Monkey GUI Testing for EDA Desktop Applications

Monkey testing is one of the autonomous Graphical User Interface (GUI) automation testing techniques. Due to the complex GUI of Electronic Design Automation (EDA) software applications, applying Monkey Testing in a timely manner is very challenging. By making use of the GUI regressions knowledge about the application we can have a faster and improved Monkey testing approach. Yet directing the improved Monkey flow to have better coverage needs enhancements so this paper addresses this by building a state graph while running to guide traversing new states using image processing and weighted random selection technique. A state is represented as an image and the improved Monkey testing flow is guided to avoid already hit windows and actions. The paper proves that the state graph-based Monkey testing approach reaches higher coverage than the improved Monkey testing approach.